Although new cancer treatment methods developed in the past decade have shown promise in minimally or noninvasive eradication of tumors, surgery remains the primary treatment paradigm for most solid tumors. Surgeons previously relied on pre-operative images for tumor resection, but recent efforts to provide image guidance in the operating room have significantly improved tumor localization. Optical imaging techniques, in particular, have found a niche in the operating room because of their relatively low cost, use of non-ionizing radiation, fast data acquisition, and real-time image guidance. However, microscopic optical technologies have small field of view, which confines their use to sampling small tissue volumes. Current intraoperative planar imaging systems display 2D images on computer monitors, requiring the surgeon to alternate between the surgical site and the monitor. The large size of these systems challenges their use in small surgical suites and hampers portability. Tomographic approaches are capable of 3D display, but the complex instrumentation and intensive image analysis precludes real-time feedback and may require steep learning curve for operators. Another related overarching problem is the prevalence of positive surgical margins because of the lack of a reliable imaging agent for intraoperative surgical margin assessment, leading to costly repeat visits. Therefore, there is a compelling need to develop an accurate, affordable, user-friendly, portable, and versatile intraoperative imaging system with real-time imaging capability. The technology platform should also be able of providing real-time assessment of surgical margins. To accomplish these goals, we will develop an intra-operative near-infrared (NIR) fluorescence/reflectance 3D imaging goggle system that can accurately image tumor boundaries, small positive nodules, and provide image guidance in real-time. Intraoperative surgical margin assessment will be aided by a tumor-selective optical imaging agent optimized for the goggle system. At the development stage, the goggle system will use information from the molecular probe to optimize the detection scheme. Specifically, we will develop and optimize a hands-free 3D head-mounted fluorescence imaging and display system, and a near infrared imaging agent for in vivo staining of tumor margins. We will then use the goggle system for molecular imaging of small animal models of breast cancer, as well as canine breast cancer patients. Pharmacology and safety data generated from these studies will be used for investigational new drug (IND) and device exemption (IDE) application to the FDA.